Editorial

Single nucleotide polymorphisms (SNPs) may introduce premature termination codons. But do they really exist? In this issue of Human Genomics, Savas et al. publish their systematic search for such SNPs (known as X-SNPs) in the dbSNP database. Surprisingly, they found 28 of them, the majority of them representing commonly occurring SNPs. Furthermore, the population specificity of these X-SNPs is clearly notable. Are they related to any diseases or traits? Not yet, but soon they will be. Also from Ozcelik's group, Savas et al. report the results of a search of functional non-synonymous SNPs (nsSNPs) in a large number of carcinogenesis-related genes expressed in breast tissue. They suggest that these nsSNPs are likely to contribute to breast cancer susceptibility.

While association analysis provides clues for the location of the genes/mutations underlying diseases, functional assays are critical to nail them. Interrogation of the impact of a missense variation on the tertiary structure of a protein may shed light on the possible functional role of the mutation of interest. Shabbeer et al. did exactly that on 29 missense mutations underlying Fabry disease, as reported in this issue.

As a special feature of this Journal, we continue to publish reviews on analytical methods. In this issue, three such reviews are published. Dudbridge et al. conducted an in-depth review of the detection of multiple associations in genome-wide studies. Motsinger and Ritchie provide an introduction to multifactor dimensionality reduction for detecting and modelling gene - gene interactions. Onkamo and Toivonen contribute a comprehensive survey on data mining methods for linkage disequilibrium mapping.

In this issue, Human Genomics publishes its first ever Letter to the Editor since it was inaugurated two years ago. I am pleased to see that this Journal is becoming an attractive forum for constructive academic discussions. More will come in the next issue.